JP2014014352A - Method for identifying stem cell state - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide useful information on a stem cell state.SOLUTION: A method of this invention for identifying a stem cell state includes: culturing stem cells each having a nucleic acid which includes a transcription factor binding site for a marker gene to detect the stem cell state and a luciferase gene located downstream of the binding site; chronologically capturing images of the luminescence due to an expression of the luciferase gene to obtain a chronological profile of the stem cells, including multiple luminous images; and analyzing a feature quantity extracted from the chronological profile.

Description

  The present invention relates to a method for identifying the state of a stem cell.

  In the field of regenerative medicine, cell transplantation using stem cells and treatment of intractable diseases by organ regeneration are expected. Although clinical applications such as heart disease have preceded, there are many unclear points about how the stem cells actually transplanted are functioning. In order to perform safe transplantation, it is essential to elucidate the mechanism of stem cell differentiation induction (induction into cells having a specific function) and to identify factors that control intracellular dynamics and functions.

  To date, stem cells have been shown to express several transcription factors that are highly specific for undifferentiated cells. These include Oct-4, Sox, Nanog, leukemia inhibitory factor receptor (LIF-R). Oct-4 is expressed in gastrulation embryos, early cleavage embryos, blastocyst inner cell mass cells, and embryonic cancer (EC) cells. In adult animals, Oct-4 is found only in germ cells.

  In stem cell experiments such as ES cells and iPS cells, pluripotency is maintained by co-culture with feeder cells and addition of LIF, but they lose their pluripotency due to the environment and cell conditions and easily differentiate. Sometimes. In order to understand the differentiation induction process, it is important to accurately know the undifferentiated state (the state having pluripotency) and the differentiated state, but it is difficult to determine the differentiated state only by the form of the stem cells.

  Patent Document 1 discloses a method including obtaining a time-lapse profile of a cell by monitoring a gene state related to at least one gene selected from genes derived from the cell over time. In the examples, the cells were fixed to a solid support, and a reporter gene such as GFP was introduced into the cells arranged in an array, and then images were acquired in time series, and the cell state was determined using a time-lapse profile. Judgment. However, in this method, quantification and evaluation by numerical analysis are performed on a cell group including a plurality of cells, and the cells cannot be individually evaluated, and it is difficult to determine the diversity of stem cells. It is. In addition, since GFP is used for observation, it is necessary to irradiate cells with excitation light. In particular, in the case of observation over a long period of time, it is necessary to consider the influence of excitation light on cells.

  Patent Document 2 extracts a region occupied by a feeder cell from an image including a stem cell colony and a feeder cell, calculates a feature amount indicating a morphological feature of the feeder cell existing in the region, and calculates the feature amount and the stem cell colony. A cell evaluation apparatus including evaluating a state of a stem cell colony based on a correspondence relationship with the state of is disclosed. However, cell culture that promotes cell differentiation may be performed without using feeder cells, and in such a case, the state of stem cells cannot be evaluated.

JP 2006-522605 A JP 2011-229409 A

  An object of the present invention is to provide useful information about the state of stem cells.

  The method for identifying the state of a stem cell according to the present invention comprises culturing a stem cell having a nucleic acid comprising a transcription factor binding region of a marker gene for detecting the state of the stem cell and a luciferase gene located downstream of the region. Imaging of luminescence due to the expression of the luciferase gene over time to obtain a temporal profile including a plurality of luminescent images, and analyzing a feature amount extracted from the temporal profile.

  According to the present invention, useful information about the state of stem cells is provided.

FIG. 1 is a diagram illustrating an example of a colony shape of each score when scoring the shape of a colony. FIG. 2 is a graph showing temporal changes in luminescence generated from (a) undifferentiated cells and (b) differentiated cells. FIG. 3 is a graph schematically showing temporal changes in gene expression levels based on differentiation markers and gene expression levels based on undifferentiation markers. FIG. 4 is a cross-sectional view schematically showing a container having a plurality of indentations on the bottom surface, which can be used for cell culture. FIG. 5 is a diagram schematically showing the state of stem cells when the container as shown in FIG. 4 is used. FIG. 6 is a diagram illustrating an example of a bright field image and a light emission image acquired in the example. FIG. 7 is a diagram showing the gene expression intensity and expression ratio of each marker in one EB. FIG. 8 is a diagram showing the gene expression intensity and expression ratio of each marker in another EB.

  1st Embodiment of this invention is related with the method of identifying the state of a stem cell.

  The method according to the first embodiment comprises culturing a stem cell having a nucleic acid comprising a transcription factor binding region of a marker gene for detecting a state of a stem cell and a luciferase gene located downstream of the region, and the luciferase This includes imaging luminescence due to gene expression over time to obtain a temporal profile including a plurality of luminescent images, and analyzing feature quantities extracted from the temporal profile.

  The stem cells are, for example, embryonic stem cells (ES cells), somatic stem cells, artificial pluripotent stem cells (iPS cells) or cancer stem cells. Stem cells may be derived from various organisms, and are stem cells from mammals such as humans and mice. The stem cell may be a commercially available cell or a cell produced by a known method.

  As the state of the stem cell, for example, the degree of differentiation, the degree of undifferentiation, and / or the degree of response to foreign factors are identified. That is, it is identified how much stem cells are differentiated and how much they respond to foreign factors. Identification of the state of the stem cell includes not only evaluating the state of the stem cell at the time of observation but also predicting the state of the stem cell after a certain time from the time of observation.

  A marker gene for detecting the state of a stem cell is a gene that makes it possible to determine the state of a cell based on, for example, whether or not the gene is expressed. The marker gene that can be used for identifying the degree of undifferentiation is, for example, a gene that is specifically expressed when cells are undifferentiated, and specific examples thereof include Nanog, Oct, Sox, and the like. Marker genes that can be used in identifying the degree of differentiation are, for example, genes that are specifically expressed in the process of cell differentiation, and specific examples thereof include neuronal differentiation marker genes such as Nestin and Mash1, and GATA4, Nkx2. 5 is a myocardial differentiation maker gene.

  A transcription factor binding region is a region present in a nucleic acid to which a protein called a transcription factor can specifically bind. The transcription of genetic information on the nucleic acid is controlled by binding of the transcription factor to the transcription factor binding region. The transcription factor binding region is, for example, a promoter, enhancer or silencer. As the transcription factor binding region, a known one may be used, or it may be cloned from a base sequence located upstream of the marker gene.

  When using a promoter as a transcription factor binding region, the promoter region of Nanog can use, for example, those described in T Kuroda et al., Molecular and Cellular Biology (2005 vol.25, No.6 p2475-2485) As the promoter region of Oct-4, for example, those described in S Okumura-Nakanishi et al., The journal of Biological Chemistry (2005 vol. 280, No7 p5307-5317) can be used, and the promoter region of Nestin is, for example, L Those described in Cheng et al., FEBS Letters 2004 565 p195-202 can be used.

  The luciferase gene is a gene encoding luciferase, which is an enzyme that catalyzes a chemical reaction in which luminescence occurs. Luciferase catalyzes the oxidation reaction of luciferin as a substrate when ATP is present. During the reaction, luciferin emits light.

  The luciferase gene may be derived from various organisms such as fireflies and bacteria. The luciferase gene may be a gene encoding a commercially available product such as Eluc luciferase that produces green luminescence, CRB luciferase that produces red luminescence, and Renilla luciferase that produces blue luminescence. Examples of commercially available vectors that already contain the luciferase gene are Eluc vector (Toyobo), CRB vector (Promega) and Renilla vector (Promega).

  The nucleic acid is, for example, DNA contained in the nucleus of a stem cell or DNA present as a vector in the cytoplasm. Part of the base sequence of the nucleic acid is a transcription factor binding region and a luciferase gene. The positional relationship between the transcription factor binding region and the luciferase gene is a positional relationship in which the expression of the luciferase gene is promoted by specifically binding the transcription factor to the transcription factor binding region.

  Only one pair of transcription factor binding region and luciferase gene may be present in the nucleic acid. Alternatively, a plurality of pairs of one transcription factor binding region and luciferase gene may be present in the nucleic acid. Alternatively, multiple types of transcription factor binding regions and luciferase gene pairs may be present one or more in the nucleic acid. When multiple pairs of transcription factor binding regions and luciferase genes are used, each pair of transcription factor binding regions is for a different marker gene, and each pair of luciferase genes results in a different color luminescence Gene.

  Stem cell culture can be performed by a known method. Stem cells may be cultured simultaneously with feeder cells. A substance for promoting or suppressing differentiation may be added to the medium for culturing. For example, Leukemia Inhibitory Factor (LIF), which is a differentiation inhibitory factor, can be added to the medium. Moreover, you may add the exogenous factor for giving a specific irritation | stimulation with respect to a stem cell to a culture medium. Stem cells are cultured, for example, at 37.0 ° C. for 1 to 3 days. Subculture can be performed as necessary. In addition, substances necessary for light emission, such as luciferin, ATP, and magnesium, can be added to the medium in an arbitrary amount at an arbitrary time based on a known method.

  Culturing can be performed using a container having a plurality of indentations on the bottom surface. FIG. 4 is a cross-sectional view schematically showing such a container. In the illustrated container 1, there are three indentations, and in each of the indentations, an embryoid body 2 composed of a plurality of stem cells is present. Although the container 1 is filled with the culture solution 3, the height A of the water surface is higher than the height (depth of the dent) B of the wall separating the dents. Each indentation may have a shape that decreases in diameter from the water surface toward the bottom surface, as shown. For example, the indentation has a U-shaped shape, a bowl-shaped shape, and the like. The maximum diameter C of the recess may be, for example, 200 μm to 3 mm.

  FIG. 5 schematically shows the state of the stem cells when the container as shown in FIG. 4 is used. As shown in FIG. 5 a, the culture solution 3 containing the stem cells 4 is added to the container 1 and left to stand. Thereafter, as shown in FIG. 5b, the stem cell 4 sinks with time and settles in each recess. Since each recess has a diameter that decreases from the water surface toward the bottom surface, the assembly of the stem cells 4 is also promoted in the recess, and finally an embryoid body 2 is formed as shown in FIG. 5c. The In addition, one stem cell can be accommodated in each recess and cultured. In this case, a culture solution whose cell concentration is adjusted so that one stem cell enters each depression is added to the container.

  Moreover, when using a container like FIG. 4, observation can also be performed using this container following culture | cultivation. That is, the observation can be performed in a state where the stem cells are contained in each recess. If the diameter of the depression is sufficiently small, a plurality of depressions can be observed within one field of view of the microscope. FIG. 6a shows a bright field image observed when using a container as in FIG. According to this figure, it can be seen that a plurality of indentations are observed within one field of view.

  In addition, as an advanced usage method of the container as shown in FIG. 4, it is possible to evaluate the effect of a specific drug. For example, after administering a drug in the culture solution, the effect of the drug and the like can be evaluated by observing the state of each recess and counting the number of recesses in which the cells are in a specific state.

  The expression of the luciferase gene is controlled according to the binding between the transcription factor binding region and the transcription factor. Since the transcription factor binding region is originally for controlling the transcription of the marker gene, the expression timing and expression level of the luciferase gene can be considered to be consistent with those of the marker gene. The luciferase protein produced in the cell by the expression of the luciferase gene catalyzes the oxidation of the luciferin provided to the cell, and as a result, the luciferin emits light. Therefore, the timing and expression level of the marker gene can be estimated based on luminescence.

  The emitted light is imaged over time and an image is acquired. In the present invention, an image obtained by imaging light emission is referred to as a light emission image. The luminescent image includes a region where a luminescent signal derived from a luminescent cell is detected and a region where a luminescent signal derived from a non-luminescent cell and a portion where no cell exists is not detected. For this reason, the light emission for every cell or every colony can be estimated.

  The luminescent image is, for example, an apparatus including a filter that mainly transmits only light having a specific wavelength corresponding to light emission, an image sensor that converts light into an electric signal, and a processing unit that generates a luminescent image from the electric signal. Can be used. An example of an apparatus for acquiring a luminescence image is a luminescence imaging system, and a specific example thereof is a luminescence imaging system LV200 (manufactured by Olympus Corporation).

  Imaging is performed over time. That is, imaging is performed continuously at an arbitrary interval. For example, imaging is performed at intervals of 5 minutes to 1 hour. In addition, the time for one imaging is arbitrarily set. One imaging time can be adjusted so that a sufficient emission signal can be detected.

  In parallel with the acquisition of the luminescent image, it is also possible to acquire a bright field image based on the light generated by the illumination light irradiated to the stem cells reflected and / or transmitted through the stem cells. That is, a bright-field image is an image that allows observation of the position and form of stem cells without being based on luminescence. The bright field image includes a phase difference image and a DIC observation image. The bright field image can be acquired at almost the same timing as the acquisition of the luminescent image, or can be acquired at any timing independent of the acquisition of the luminescent image.

  A temporal profile means a collection of information on stem cells obtained by measuring stem cells over time. The temporal profile includes a plurality of luminescent images acquired over time. The temporal profile may also include one or more bright field images. Furthermore, the temporal profile may include information other than images, and may include, for example, conditions during culture such as culture time, culture temperature, and culture solution composition.

  A feature amount is extracted from the temporal profile. The feature amount is information obtained by quantifying a part of information about the stem cell included in the temporal profile, presence / absence of a certain feature, variation information such as a sign change, and the like.

  An example of the feature amount is an amount based on light emission extracted from the light emission image. Examples of amounts based on luminescence are the intensity of luminescence and the variation in luminescence. The intensity of light emission is, for example, the absolute or relative intensity of light emission at a specific time, or the intensity of light emission integrated or averaged over a specific time period. The fluctuation of the luminescence can be found, for example, from a curve obtained by plotting a measured value of luminescence intensity on a plane of luminescence intensity-measurement time and connecting each plot. For example, the variation in light emission is the shape of such a curve, the position of the peak of light emission intensity in such a curve, and the like. Further, when at least a part of such a curve vibrates, it is information relating to the vibration, for example, the wavelength or period, amplitude or wave height of the vibration. Vibration means repeated increase / decrease in the amount of light emission.

  As another example of the feature amount, an amount based on the form of the stem cell, that is, the form of the stem cell is digitized based on an arbitrary criterion. An amount based on the morphology of the stem cells can be extracted from the luminescence image and / or the bright field image. Examples of quantities based on stem cell morphology are stem cell colony thickness, stem cell circularity, stem cell size, and stem cell colony size.

  The extracted feature quantity is then analyzed, but the feature quantity can be mathematically processed before the analysis. By this processing, the feature amount is converted into one suitable for analysis. An example of such mathematical processing is to convert each of the feature values into an arithmetic average between at least one feature value acquired before and after the feature value. Alternatively, each feature amount is converted into a variance value.

  The analysis is performed, for example, to determine the state of the stem cell. This determination is to evaluate the current state of the stem cell or to predict the future state of the stem cell.

  The analysis can be performed based on a known method. For example, when a plurality of types of feature values are extracted, the feature values can be subjected to multivariate analysis as analysis. Examples of multivariate analysis are discriminant analysis, logistic regression analysis, principal component analysis or factor analysis. Discriminant analysis or logistic regression analysis can be performed based on a linear model or a non-linear model. Alternatively, the autocorrelation coefficient may be used as a variable. Various analyzes can be performed simultaneously.

  According to the method according to the first embodiment, the state of the stem cell can be identified with high accuracy by analyzing based on the feature amount extracted from the temporal profile. In particular, since luminescence caused by luciferase is used, damage to cells can be minimized, and long-term observation over several days to several weeks becomes possible. Moreover, since a luminescent image is acquired, a state can be identified for every stem cell and every colony. Therefore, it is possible to identify the state of the stem cell with higher accuracy as compared with the conventional method. As a result, application of stem cells to regenerative medicine and the like can be promoted, and elucidation of stem cell pluripotency maintenance mechanisms, differentiation mechanisms, and the like can be promoted. If the differentiation state of stem cells in individual stem cells or stem cell colonies can be identified, stem cells or stem cell colonies suitable for the purpose can be selected and recovered and used appropriately when stem cells are used for cell therapy or research.

  For cell culture, generally, a plate is used in which each well is separated and its diameter is about 1 cm. In such a plate, since the culture solution does not move between the wells, the environment may be different for each well. Further, when observing using a microscope, it is difficult to capture the state of a plurality of wells in one field of view. Furthermore, in a plate having a flat bottom surface, cells are also easily moved by convection of the culture solution. Since convection of the culture medium is easily caused by changes in the ambient temperature or moving the plate, for example, such a plate is not suitable for observations that avoid moving cells as much as possible, such as observation over time. It becomes.

  On the other hand, in a container having a plurality of depressions on the bottom surface as shown in FIG. 4, the water surface of the culture solution is higher than the height of the wall separating the depressions, so that the cells present in each depression are in the same environment, They can be cultured separately from each other. In addition, when a microscope is used, a plurality of indentations can be observed at the same time, so that it is easy to check the variation in the state of the cells, and the reliability of the observation results is improved. In addition, because the cells are contained in each well, even if there is some convection in the culture or even if the cell itself moves, There is little to move into the indentation. Therefore, it is suitable for use for observation over time.

  The second embodiment of the present invention relates to a method for identifying the state of a stem cell.

  The method according to the second embodiment includes the method according to the first embodiment including predicting the state of the stem cell at a certain time in the future and / or during a certain period based on the result of analyzing the feature amount. It is.

  A future point in time and / or period is, for example, a specific point in time and / or period when a certain period of time has elapsed after the acquisition of a time-lapse profile has been completed, or constant after the method has been completed. A specific point in time and / or duration when time has passed.

  The prediction can be performed based on, for example, the result of analyzing the feature amount. For example, when discriminant analysis or logistic regression analysis is performed based on the feature amount, the state of the stem cell at a certain future time point or period is identified based on the calculated discriminant function or prediction formula.

  The stem cell whose state is to be predicted does not have to be the same as the stem cell used to extract the feature amount. For example, after culturing a small amount of stem cells, extracting feature values from the stem cells, obtaining the analysis results in advance, and then applying the analysis results to stem cells cultured under almost the same conditions to determine the future state. Can be predicted.

  According to the method according to the second embodiment, the future state of the stem cell can be predicted with high accuracy. Thereby, for example, when applying stem cells to regenerative medicine, the success rate of cell transplantation and organ formation can be improved.

  The third embodiment of the present invention relates to a method for diagnosing a subject.

This method
a) monitoring the state of transcription associated with at least one of the transcription factor binding regions inherent in the cell over time to obtain a temporal profile;
b) determining the state of the cell based on the time-lapse profile; and c) determining the state, disorder or disease of the subject from the state of the cell.

  In this method, the cells may be collected from a subject. The method may further include applying to the subject a treatment or prevention selected according to the determined result.

  The fourth embodiment of the present invention relates to a system for diagnosing a subject.

This system
a) means for monitoring the state of transcription associated with at least one of the transcription factor binding regions inherent in the cell over time to obtain a temporal profile;
b) means for determining the state of the cell based on the temporal profile; and c) means for determining the state, disorder or disease of the subject from the state of the cell.

  In this system, the cells may be taken from a subject. The system may further include means for providing the subject with a treatment or prevention selected according to the determined result.

  The method according to the third embodiment and the system according to the fourth embodiment include drug resistance, screening for low molecular weight compounds that promote pluripotency maintenance and differentiation induction, selection of an appropriate anticancer agent, and appropriate transplanted cells. It is applicable to tailor-made diagnosis and treatment such as selection.

<Example 1: Evaluation of stem cell status using undifferentiated marker Nanog gene>
The Nanog gene is widely used as an undifferentiated marker for stem cells, but the flow cytometer analysis results show that gene expression varies over time, and the degree of differentiation can be determined simply by examining transient gene expression levels. It has been reported that it is difficult to do (Chambers et al., Vol 450, p1230, Nature 2007).

  Based on the present invention, the state of stem cells was evaluated. As a transcription factor binding region, a transcription factor binding region for controlling transcription of the Nanog gene, which is an undifferentiated marker, was used. From the luminescence image, the feature amount related to the fluctuation of the expression of the Nanog gene was extracted. Furthermore, the feature-value regarding the shape of a cell was extracted from the bright field image.

(1-1) Preparation of cells A cell having a nucleic acid containing a promoter for the Nanog gene, which is an undifferentiation marker, and a luciferase gene located downstream thereof in the nucleus was prepared. As its outline, a gene transfer vector containing a transcription factor binding region, a luciferase gene, and a drug resistance gene was introduced into a stem cell, and a cell in which the drug resistance gene was stably expressed (stable expression cell) was selected. Details are described below.

  The promoter for the Nanog gene was cloned with reference to known literature (T Kuroda et al., Molecular and Cellular Biology 2005 vol. 25, No. 6, p2475-2485). Specifically, the following primers were used using mouse-derived genomic DNA as a template.

forward primer: CTACTCGAGATCGCCAGGGTCTGGA (SEQ ID NO: 1)
reverse primer: CTACTCGAGCGGCAGCCTTCCCACAGAAA (SEQ ID NO: 2).

  On the other hand, the nucleic acid fragment corresponding to the luciferase gene in the neomycin resistant pGL4 vector (Promega) was replaced with a fragment of the Eluc luciferase gene. A fragment corresponding to the promoter of the Nanog gene cloned as described above was inserted into this vector. The gene transfer vector thus obtained is referred to as Nanog gene expression specific luminescence vector pNanog-Eluc.

  In order to transfect pNanog-Eluc into mouse ES cells, ES cells were first cultured with feeder cells. Specifically, a 35 mm diameter plastic dish was coated with 0.1% gelatin and washed 3 times with PBS. MEF cells (mouse embryonic fibro-blast mouse embryo fibroblasts) whose division was stopped by mitomycin C treatment were seeded as feeder cells in a gelatin-coated dish, and cultured overnight. As the medium, DMEM (with phenol red and 10% FCS) was used. On the next day, mouse ES cells (BRC6 strain, RIKEN BRC) were plated on feeder cells in a 35 mm dish and cultured overnight.

  Subsequently, pNanog-Eluc was transfected into mouse ES cells. Transfection was performed by Nucleofection method using Amaxa Nucleofector (Wako Pure Chemical Industries, Ltd.). An appropriate amount of G418 was added to the medium, and only cells into which the gene was stably introduced were selected.

(1-2) Cell Culture Mouse ES cells constitutively expressing Nanog-Eluc were cultured overnight in DMEM medium supplemented with 15% KSR (Knockout Serum (Gibco)) and LIF (leukemia inhibitory factor). did. Thereafter, the cells were divided into two, and each was added to DMEM medium (15% KSR, phenol red-excluded) supplemented with LIF and HEPES and DMEM medium (15% KSR, phenol red-excluded, LIF-free) supplemented with HEPES, respectively. Moved. As LIF, the product names LIF Human, recombinant, Culture Supernatant (manufactured by Wako Pure Chemical Industries, Ltd.) were used at specified concentrations.

(1-3) Acquisition of Luminescent Image After adding D-luciferin (manufactured by Promega Corporation) with a final concentration of 500 μM to the medium, the medium was installed in a luminescence imaging system LV200 (manufactured by Olympus Corporation). Mouse ES cells were imaged for 12 minutes at 15 minute intervals, the objective lens used was a magnification of × 20, the binning was 1 × 1, and the CCD camera was ImagEM (manufactured by Hamamatsu Photonics).

  The acquired luminescence image was analyzed by numerical data using an image analysis system AQUACOSMOS (manufactured by Hamamatsu Photonics) and graphed. From the luminescence image of the cells cultured in the absence of LIF, the luminescence information was quantified by classifying into undifferentiated colonies and slightly expanded colonies using morphological information from bright field images. A cell selection region of Region of interest (ROI), each containing one colony, was selected. Undifferentiated cells were selected as ROI6-10 and differentiated cells were selected as ROI1-5.

  FIG. 2A shows the light emission amount with respect to time for the ROIs 6 to 10, and FIG. 2B shows the light emission amount with respect to time for the ROIs 1 to 5. 2 (a) and 2 (b), the expression level of the Nanog gene is higher in cells that are morphologically undifferentiated (FIG. 2a) than in cells that are morphologically differentiated (FIG. 2b). I understand that it is expensive. Furthermore, according to FIG. 2 (a), it can be seen that the expression level of the Nanog gene oscillates with the progress of time in cells that are morphologically undifferentiated.

(1-4) Scoring about the form of the cell based on a bright field image The bright field image was acquired simultaneously with acquisition of the light emission image. In the bright field image, ROIs each containing one colony were designated to correspond to those in the luminescent image, and the degree of differentiation was judged from the shape of the colonies contained in each ROI.

  The degree of differentiation was classified into three stages shown in FIG. 1, and three scores were assigned to each stage. As the most undifferentiated stage, score 2 is a colony with a compact sphere and clear edge, and as a middle level, score 1 is a colony at a stage where the thickness has decreased and the sphere has started to collapse slightly, As the most differentiated stage, a colony that had collapsed and became flat was scored as 0. These determinations were made visually. In addition, since it was thought that distributing colonies in the middle of a completely undifferentiated stage and a fully differentiated stage to either of these two stages would reduce the accuracy of the analysis results, a medium stage was provided. . This determination may be made automatically using an image analysis program or the like.

  The colonies included in each ROI were scored for the colony shape at regular intervals. The scoring results for one of the ROIs are shown in Table 1 below. Table 1 summarizes the results of scoring the colony shape over time and the relative values of the amount of luminescence (that is, the amount of gene expression).

  From Table 1, it can be seen that the score representing the shape decreases as time progresses. This shows that this colony differentiates with time progress. In contrast, it can be seen that the gene expression level fluctuates and fluctuates. Comparing two arbitrary time points with the same shape-representing score, it can be seen that the gene expression level may be different. Moreover, when arbitrary two time points where the gene expression level is the same are compared, it can be seen that the score representing the shape may be different.

(1-5) Discriminant analysis From the relationship between the light emission amount and the cell shape as described above, even if the cell shape is the same, the cell in which the expression level of the Nanog gene oscillates finally differentiates. It has been confirmed that there is a tendency for the time to be longer.

  The sample was divided into two groups in the morphological differentiation state after 53.5 hours, and the two groups were compared by the Mann-Whitney U test with the number of vibrations up to 35 hours. It was found that the number of vibrations was significantly higher (p = 0.0152).

  Therefore, a discriminant analysis for discriminating the differentiation state was performed using the degree of oscillation of the gene expression level and the cell shape score as variables.

  For each ROI, the gene expression level was measured every hour, and the number of vibrations was calculated from the measurement result. Specifically, the gene expression level was measured for each time point, and the process of smoothing those values was performed. That is, three measured values before and after the measured value at each time point were added, and an average value was calculated from the measured values for a total of seven points. This processing was performed for the purpose of reducing the influence of noise included in the measurement result at that time. Next, in the value after smoothing at each time point, the amount of change (increase / decrease) at two consecutive time points was calculated, and the number of times the sign of the increase / decrease was reversed was defined as the number of vibrations. For each ROI, the vibration in the gene expression level from the start of measurement to 19.5 hours later was calculated.

  On the other hand, for each ROI, a colony shape score at the time when 19.5 hours elapsed from the start of measurement was calculated.

  Further, for each ROI, it was determined whether or not it was differentiated when 53.5 hours had passed. Judgment results were expressed as numbers. That is, 0 was set when differentiated, and 1 was set when undifferentiated.

  Nine ROIs, shape score at 19.5 hours (x), number of oscillations of gene expression up to 19.5 hours (z), and whether differentiation is at 53.5 hours Table 2 below summarizes the determination results.

  Furthermore, the shape score (x) at the time of 19.5 hours, the number of oscillations of the gene expression level until 19.5 hours (z), and the determination result of whether or not the differentiation is at the time of 53.5 hours Based on this, a discriminant function W for predicting differentiation at the time point of 53.5 hours was obtained as follows.

  In this equation, when W is a positive value, the cell is undifferentiated, and when W is a negative value, the cell is differentiated.

  In the rightmost column of Table 2, the value of W calculated by substituting the x and z values of each ROI is shown. When this value was compared with the actually observed state of differentiation, 8 of the 9 ROIs matched. In other words, the hit rate was 88.89%.

  For comparison, when predicted based on gene expression level alone, only 4 out of 9 ROIs were hit (Hit rate = 44.44%), and when predicted based only on colony shape, Only 5 of the ROIs were hit (Hit = 66.67%).

  Moreover, when the state of differentiation after 46.5 hours was predicted based only on the frequency of gene expression level oscillation from the 10-hour time point to the 30-hour time point, 9 out of 9 ROIs were relevant (target) (Medium rate = 100.0%). On the other hand, when predicting the state of differentiation after 46.5 hours based only on the shape score at 30 hours, 5 of the 9 ROIs were hit (target = 66.67%).

  From the above, it was found that a highly accurate prediction can be made by selecting a variation in gene expression level and a score for the shape as a characteristic amount and performing discriminant analysis based on the score. It was also found that the accuracy of prediction can be increased by selecting an appropriate range.

(1-6) Logistic regression analysis Using the measurement result of gene expression level based on luminescence image and the score related to cell shape based on bright-field image, examining the differentiation state of cells by logistic regression analysis did.

  That is, the shape x (tm) of a cell that has not yet fully differentiated (x = 0) after tm time, and the expression data y (t) of Nanog gene from ts time to tm time (t = ts to tm) ) Is used to predict the probability P (tn) that the shape of the cell after tn (tn> tm) time is in a fully differentiated state. It should be noted that once fully differentiated cells do not return to an undifferentiated state.

  When the prediction formula is expressed by a linear model, it is expressed by the following formula.

  However, the prediction formula does not necessarily need to be a linear model, and may be a nonlinear model.

  In addition, in order to add information that the expression level of the Nanog gene is oscillating, when the number of changes in the expression level of the Nanog gene from the time ts to the time tm (number of vibrations) z is incorporated into the model, The prediction formula is expressed by the following formula.

  Also in this case, the prediction formula is not necessarily a linear model, and may be a non-linear model.

<Example 2: Evaluation of stem cell state when exogenous factor is administered>
It is known that differentiation is induced when fibroblast growth factor (FGF) is administered to stem cells. On the other hand, it has been found that by administering an inhibitor of FGF to a cell in which differentiation is induced in this manner, the cell attempts to return to an undifferentiated state and is slightly differentiated.

  Thus, when using foreign factors, such as FGF and its inhibitor, the data used for analysis are selected suitably. For example, when only differentiation induction by FGF is performed, time series data after induction can be used, and when an inhibitor is used, time series data after use can be used.

  From the time series observation value of the gene expression change and the cell shape change time series observation value of the ES cell after induction of differentiation by FGF or after using the inhibitor, the differentiation state of the cell after a certain time, and within a certain time period after the certain time The differentiation state is predicted. The prediction accuracy can be improved by using the temporal profile of the shape change.

<Example 3: Evaluation of the state of a stem cell when multiple types of marker genes are used>
Various types of transcription factor binding regions and luciferase gene pairs can be used in combination to assess the degree of stem cell differentiation. By combining them, the accuracy can be improved.

  For example, a differentiation marker gene and an undifferentiation marker gene can be used in combination. By distinguishing and labeling multiple types of marker genes with a plurality of luciferase genes having different wavelength characteristics, and performing imaging for each wavelength specific to each luciferase using a spectral filter, as shown in FIG. Temporal change profile data of the light emission amount based on the differentiation marker and the light emission amount based on the undifferentiation marker is obtained.

  For example, when the ratio of the differentiation marker gene expression level to the undifferentiation marker (differentiation marker expression level / undifferentiation marker expression level) is used as the feature amount, this ratio increases with time as the stem cell differentiation proceeds. The state of the stem cell can be identified as an index.

  Further, a plurality of types (for example, two or three types) of undifferentiated marker genes may be used, one type of undifferentiated marker gene and one type of differentiated marker gene may be used, or one type of undifferentiated marker gene may be used. Differentiation marker genes and multiple types (for example, two or three types) of differentiation marker genes may be used.

  Examples using two types of combinations of transcription factor binding regions of marker genes and luciferase genes are shown below.

<Observation of ES cells before differentiation induction>
For detection of the expression of the undifferentiation marker Nanog gene and the neuronal differentiation marker Nestin gene, Eluc luciferase (observed in green) and CBR luciferase (observed in red) with different wavelength characteristics were used. When stem cells were induced to differentiate into the nervous system, changes in the expression of each marker gene accompanying differentiation induction of ES cells were detected by luminescence imaging and quantified.

  [1: Preparation of ES cell into which both the promoter region of an undifferentiated marker gene and a luciferase gene and the fusion gene of a promoter region of a differentiation marker gene and a luciferase gene are introduced]

  Cloning of the promoter region of the Nanog gene and the Nestin gene was performed with reference to the gene sequences already published in the paper. Regarding the Nanog gene, non-patent literature “T Kuroda et al., Molecular and Cellular Biology 2005 vol.25, No.6, p2475-2485” is referred to, and for the Nestin gene, non-patent literature “L Cheng et al., FEBS”. Promoter sequences were obtained with reference to Letters 2004 565 p195-202.

The sequence of the promoter region of Nanog gene was amplified using mouse genomic DNA as a template and using the following primer set.
forward primer: CTACTCGAGATCGCCAGGGTCTGGA (SEQ ID NO: 1)
reverse primer: CTACTCGAGCGGCAGCTTCCCCACAGAAA (SEQ ID NO: 2)

The sequence of the promoter region of the Nestin gene was amplified using mouse genomic DNA as a template and using the following primer set.
forward primer: GAGAACGCGTGGGCTGTGTGTGTCACT (SEQ ID NO: 3)
reverse primer: GAGACTCGAGGTGGAGCATAGAGAGAGGGAGT (SEQ ID NO: 4)

  The promoter sequences of the obtained Nanog and Nestin genes were incorporated into ELuc vector (Toyobo) and CBR vector (Promega), respectively, and “undifferentiated marker expression specific luminescent vector pNanog-Eluc” and “differentiation marker expression” A specific luminescent vector pNestin-CBR ”was prepared.

  Feeder cells were prepared for culturing ES cells for gene transfer. Specifically, a 35 mm diameter plastic dish was coated with 0.1% gelatin and washed three times with PBS. Mouse embryo fibroblast (MEF) cells whose division was stopped by mitomycin C treatment were seeded on the gelatin-coated dish, and cultured overnight. As a medium, DMEM (with phenol red and 10% FCS) was used.

  The next day, mouse ES cells (BRC6 strain, RIKEN BRC) were seeded on feeder cells in a 35 mm dish. After overnight culture, both pNanog-Eluc vector and pNestin-CBR vector were transfected into mouse ES cells and the transfected cells were treated with 15% Knockout Serum (KSR) (Gibco) and leukemia inhibitory factor (LIF). The added DMEM was used as a medium and cultured overnight. For gene transfection, the Nucleofection method by Amaxa Nucleofecor (Wako Pure Chemical Industries, Ltd.) was used. The next day, the medium was replaced with DMEM containing HEPES (15% KSR, without phenol red), and 1 μM Retinic acid (Sigma) was added to induce differentiation into neural cells.

  [2: Observation and analysis of ES cells after differentiation induction]

  When inducing differentiation of ES cells, it is common to form embryoid bodies (EB) in a suspension culture system, and a plurality of production methods such as a hanging drop method are known. In this example, A container as shown in FIG. 4 was used. That is, a special phospholipid polymer is coated on the culture surface, and an embryoid body is formed using EZ Sphere (IWAKI) having a well-like structure (microwell structure) having a U-shaped bottom in the dish, It was detected how Nanog expression and Nestin gene expression in the differentiated embryoid body changed.

Preparation of embryoid bodies (EB) was performed as follows.
(1) Confluent mouse ES cells were prepared for one 35 mm diameter dish.
(2) After washing once with PBS, trypsinization was performed.
(3) The number of cells was adjusted to 0.2 × 10 ⁵ cells / well in the end, and gene transfer was performed by Nucleofection.
(4) Cells were seeded in DMEM (15% KSR, phenol red-excluded) with luminescence observation medium HEPES added with D-Luciferin (final concentration 500 μM), and EZ Sphere (EZ Sphere, AGC Asahi Techno Glass Co., Ltd.) ), And left standing in LUMINOVIEW (LV200, Olympus Corporation) to observe light emission. EZ Sphere used a 500 μm well diameter type. An LV200 (Olympus Corporation) is used as an observation device, a 10 × objective lens (UPlanFLN 10x NA0.30) is used as an objective lens, an ImagEM (Hamamatsu Photonics Corporation) is used as a CCD camera, and EM gain is used. : 1200 observation.
(5) The culture start date was defined as day 0 of embryoid body (EB) formation. FIG. 5 schematically shows how ES cells are seeded (FIG. 5a) until ES cells are assembled (FIG. 5b) and embryoid bodies (EB) are formed (FIG. 5c).

  Observation images taken continuously from the start of embryoid body (EB) formation until 88 hours were stored, and numerical data analysis was performed using the image analysis system AQUACOSMOS (Hamamatsu Photonics Co., Ltd.).

  FIG. 6 is an example of the acquired bright-field image (FIG. 6a) and emission image (FIGS. 6b and 6c). According to FIG. 6a, the appearance of one EB in each indentation can be observed in one field of view. FIGS. 6b and 6c show the light emission of Eluc luciferase and the light emission of CBR luciferase, respectively, in the portion surrounded by the square line in FIG. 6a.

  Based on the emission intensity obtained for each spectral filter, the emission intensity derived from each luciferase is calculated based on the emission intensity obtained for each spectral filter using the calculation method used in conventional fluorescence observation. The gene expression intensity and expression ratio of were examined. The light emission of Eluc luciferase reflecting Nanog expression is spectroscopically analyzed using filter BP515-560, and the light emission of CBR luciferase reflecting Nestin expression is spectroscopically processed using filter 610ALP. It was.

  7 and 8 show the gene expression intensity and expression ratio of each marker in one EB, respectively.

  FIG. 7 shows an observation result of an embryoid body formed in the microwell selected as the region of interest 1 (ROI-1). According to the bright-field image, it was observed that the ROI-1 embryoid body was more differentiated. On the other hand, FIG. 8 shows an observation result of an embryoid body formed in the microwell selected as the region of interest 7 (ROI-7). According to the bright field image, it was observed that the ROI-7 embryoid body did not proceed with differentiation.

  FIG. 7a and FIG. 8a show images of Eluc luciferase luminescence (that reflects the state of Nanog expression) and CBR luciferase luminescence (ie, state of Nestin expression) at 0 hours after embryoid body formation, respectively. Reflected) images and those images after 88 hours are shown. FIG. 7b and FIG. 8b show the relationship between the passage of time and the amount of luminescence of Eluc luciferase (ROI_1_nanog) and CBR luciferase (ROI_1_nestin), respectively. Furthermore, FIGS. 7c and 8c show the relationship between the passage of time and the ratio of CBR luciferase to Eluc luciferase, respectively.

  In the ROI-1 embryoid body, it can be seen from FIG. 7a that the emission of ELuc luciferase reflecting Nanog expression was intensely emitted at 0 hours, but weakened at 88 hours. On the other hand, regarding the luminescence of CBR luciferase reflecting Nestin expression, it is found that the luminescence is weak at 0 hours, but slightly increased at 88 hours. This is consistent with the result of FIG. That is, although the amount of luminescence of ELuc luciferase is high at the beginning of observation, the amount of luminescence decreases with the passage of time, while the amount of luminescence of CBR luciferase slightly increases with the passage of time. From FIG. 7c it can be seen that these ratios increase with time. From the above, it was found that Nanog expression decreased and Nestin expression increased as differentiation induction progressed over time.

  On the other hand, in ROI-7 embryoid bodies, it can be seen from FIG. 8a that the luminescence observed at 0 hours for ELuc luciferase is maintained to some extent at 88 hours, and that for CBR luciferase is 0 hours and It can be seen that light emission is weak at both time points of 88 hours. This agrees with the result of FIG. 8b, and the amount of luminescence of ELuc luciferase gradually decreases with time, and the amount of luminescence of CBR luciferase repeatedly decreases and increases in the measured range. . From FIG. 8c, it can be seen that these ratios are maintained within a certain range with some increase or decrease. From the above, it was found that ROI-7 does not show the same expression as ROI-1 even when cultured under differentiation conditions.

  When two types of luminescent genes are used as in this example, the increase or decrease of one luminescent gene may be used in the evaluation of the observation result, or the increase or decrease of both luminescent genes may be used. In particular, in the latter case, a ratio of two luminescent genes may be used. Furthermore, when using a ratio, the value of the ratio may be used directly, or the amount of change in the ratio up to the elapse of a certain time may be used.

  Further, by performing continuous observation on the same embryoid body rather than comparing the expression levels temporarily, the gene expression level or the ratio thereof can be detected as a profile, and the accuracy can be improved.

  According to this example, it was shown that the degree of differentiation and directionality may differ depending on each embryoid body even if the culture is performed under the same culture medium. In this way, the degree of differentiation varies from embryoid body to embryoid body, so that it is suitable for subsequent experiments by culturing and observing multiple embryoid bodies within one field of view under the same culture conditions. It becomes possible to select embryoid bodies in a differentiated state.

<Example 4: Evaluation of homogeneity of each stem cell or each stem cell colony>
In recent years, research for differentiating iPS cells and ES cells into target cells and using them for cell therapy has been accelerated, and selection of appropriate stem cell lines (stem cell clones) has been advanced. However, according to Development 135, 909-918 (2008), ES cells are a heterogeneous population of cells with a mixture of inner cell mass-like cells and early primitive ectoderm-like cells that are mutually convertible. It is reported that when LIF is removed from the medium, the former differentiates into cells of extraembryonic tissue and the latter tends to differentiate into somatic cells. Thus, it is known that the properties of stem cells easily change depending on the environment and cell conditions. In addition, by culturing for a long period of time, stem cells that lose pluripotency and become differentiated may be mixed in the stem cell population, and there is a concern that differentiation pluripotency may vary.

  On the other hand, when iPS cells or ES cells are differentiated into target cells and used for cell therapy, if the cells to be transplanted contain undifferentiated cells, there is a risk that they will become a tumor after transplantation, and they will be completely differentiated. Cells must be used for cell therapy.

  Thus, it is important to grasp the homogeneity of the differentiation state of individual stem cells in the cell population of the stem cell line.

  As an example of analysis, time-lapse profile data of the expression level of Nanog gene, which is a pluripotency marker, is obtained, feature quantities such as the number of vibrations are extracted, the differentiation state in individual stem cells or stem cell colonies is examined, and the cell group The homogeneity of the feature quantity can be evaluated. The number of stem cells or colonies is preferably about several tens to several hundreds. For evaluation of homogeneity, time-lapse profile data of different cell groups to be compared may be used. For example, in order to evaluate the influence of the number of passages of stem cells on the differentiation state, time-lapse profiles of expression of undifferentiated markers such as Nanog genes were compared between stem cell lines with low passage numbers and stem cell lines with high passage numbers. In some cases, cells with a low passage number will have a higher homogeneity in the Nanog expression profile, and homogeneity will decrease as the number of passages increases.

  As an analysis method, for example, techniques such as outlier test, chi-square test, and cluster analysis can be used.

  By evaluating the homogeneity of the iPS cell group or ES cell group, the quality evaluation and management of stem cells for the purpose of cell therapy or research use are performed.

Claims (16)

Culturing a stem cell having a nucleic acid comprising a transcription factor binding region of a marker gene for detecting the state of the stem cell and a luciferase gene located downstream of the region;
Imaging luminescence due to expression of the luciferase gene over time, obtaining a temporal profile including a plurality of luminescence images;
A method for identifying a state of a stem cell, comprising analyzing a feature amount extracted from the temporal profile.   The method according to claim 1, wherein the feature amount includes an amount based on a shape of the stem cell.   The amount based on the morphology is at least one selected from the group consisting of a thickness of the stem cell colony, a roundness of the stem cell, a size of the stem cell, and a size of the stem cell colony. The method described.   The method according to claim 1, wherein the feature amount includes an intensity and / or variation of the light emission.   The method according to any one of claims 1 to 4, wherein the stem cells are ES cells, somatic stem cells, iPS cells, or cancer stem cells.   The method according to any one of claims 1 to 5, wherein as the state of the stem cell, the degree of differentiation, the degree of undifferentiation, and / or the degree of response to a foreign factor are identified.   The method according to any one of claims 1 to 6, wherein the transcription factor binding region is a promoter, an enhancer, or a silencer.   The nucleic acid includes two or more combinations of the transcription factor binding region and the luciferase gene, and the transcription factor binding regions and the luciferase genes included in different combinations are different from each other. The method according to item. The culture is performed using a container having a plurality of depressions on the bottom surface, and the culture solution contained in the container is accommodated in the container such that the water surface is higher than the height of the wall separating the depressions. And
The method according to any one of claims 1 to 8, wherein the imaging is performed in a state where the stem cells are accommodated in each of the recesses.   The method according to claim 9, wherein the maximum diameter of the indentation is 200 μm to 1 mm.   The method according to any one of claims 1 to 10, further comprising predicting a state of the stem cell at a future time point and / or a period based on a result of analyzing the feature amount.   12. The method according to claim 1, wherein a plurality of types of feature quantities are extracted, and the analysis includes multivariate analysis of the plurality of types of feature quantities.   The method according to claim 12, wherein the multivariate analysis is discriminant analysis or logistic regression analysis.   The method according to claim 13, wherein the discriminant analysis or logistic regression analysis is performed based on a linear model or a nonlinear model, or is performed using an autocorrelation coefficient as a variable.   15. The method according to any one of claims 1 to 14, further comprising: mathematically processing the feature amount before the analysis to convert it into a suitable one for the analysis.   The method according to claim 15, wherein the mathematical process is converting each of the feature quantities into an arithmetic average between at least one feature quantity acquired before and after the feature quantities.

Images